# The Brief That Changed My Job Title > I wanted simple wireframes. External users loved the busy version instead. That surprise changed what we were actually building. - Author: Alejandro Haydar - Published: 2024-12-15 - Category: Product Design - Canonical: https://alejandroastroport.netlify.app/case-studies/cva-01-the-brief-that-changed-my-job-title/ --- I wanted to keep the wireframes simple. This was late 2024. Thought Industries — the learning management platform where I led design — had an idea leadership was excited about. An AI system that would help companies understand how the market actually perceived their value. Not through internal surveys or analyst reports. Through real signals: G2 reviews, job postings, competitor content, social mentions. A weekly score. A feed of market intelligence. My instinct was to strip it down. One metric. Clean layout. Minimal decisions per screen. That's usually the right call in a first wireframe — you want to see if the core idea holds before you load it up with information. Leadership pushed for more. They wanted the data surfaced. All of it. I went with their version. --- ## We Were Not Building for the Users We Already Had Before I get to the research, there's context that matters: this wasn't a product for Thought Industries' existing customers. Thought Industries runs learning platforms for B2B companies — software companies, manufacturers, financial services firms — that want to train their external audiences. Their existing customers were already bought in. They understood the product. This was aimed at a different buyer entirely. Customer education leaders at companies that *own* an LMS but couldn't articulate the business case for it internally. People who ran customer training programs and struggled to connect that work to revenue, retention, or competitive positioning. It was a bet on a new ICP. And that bet shaped everything about how we'd test. --- ## The Research Setup We didn't test internally. We used a research team to recruit external potential customers — people fitting the new ICP, not Thought Industries users. Real market discovery. The question we were trying to answer wasn't "do users understand this interface?" It was: do people in this role believe this product would change anything about how they do their job? We ran concept tests with both versions. The clean one. The busy one. --- ## What the Research Came Back With The busy version won. Not narrowly. Users looked at the information-rich wireframe — competitor breakdowns, signal categories, a weekly score with underlying data — and said something I didn't expect. Not "this is a lot" or "I don't know where to look." They said: *finally, something that sees the full picture.* They'd been working with tools that gave them fragments. One platform tracked reviews. Another scraped job postings. They were manually synthesizing it themselves, or not doing it at all. The density we'd designed wasn't overwhelming to them — it was the first time they'd seen all of it in one place. My instinct to simplify had been wrong. The users weren't asking for less. They were asking to be trusted with more. --- ## What That Reframed When you're wrong in research, you have two choices: argue with the data or update your understanding of the problem. The data was clear. So the problem changed. The challenge wasn't complexity reduction. The challenge was *activation*. We had users who wanted the richness. But wanting to see the picture and knowing what to do with it are different problems. That's a harder product to build. And it planted a question that stayed with me for the next eighteen months: when you're building a system powered by decisions the user doesn't get to see — an algorithm, a model, a framework operating under the surface — what do you owe them? What should be visible, and what should stay hidden? I didn't have an answer yet. But the wireframes had just proven the product concept was real. Leadership wanted it to move — to feel real, to show the AI in action, not just describe it. Mid-summer, they handed me GitHub access. --- *This is part one of a four-part series on building the Customer Value Agent at Thought Industries. [Part two: The Swipe That Didn't Work.](/feed/cva-02-the-swipe-that-didnt-work/)*